On 2019-01-22 5:40 a.m., Greg Kroah-Hartman wrote:
> On Wed, Jan 16, 2019 at 11:25:17AM -0700, Logan Gunthorpe wrote:
>> This is resend number 6 since the last change to this series.
>>
>> This cleanup was requested by Greg KH back in June of 2017. I've resent the
>
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
hus I've
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
--
Logan Gunthorpe (6):
iomap: Use non-raw io functions for io{read|write}XXbe
parisc
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
x27;ve
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
--
Logan Gunthorpe (6):
iomap: Use non-raw io functions for io{read|write}XXbe
parisc: i
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
e not defined when iomap.h was included. Thus I've
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
Logan Gunthorpe (6):
iomap: Use non-raw io functi
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
ticed by Horia.
(earlier versions were drastically different)
--
Logan Gunthorpe (6):
iomap: Use non-raw io functions for io{read|write}XXbe
parisc: iomap: introduce io{read|write}64
iomap: introduce io{read|write}64_{lo_hi|hi_lo}
io-64-nonatomic: add io{read|write}64[be]{_lo_hi|_hi_lo} m
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
writeq were not defined when iomap.h was included. Thus I've
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
--
Logan Gunthorpe (6):
iomap: Use non-
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
*** BLURB HERE ***
Logan Gunthorpe (6):
iomap: Use non-raw io functions for io{read|write}XXbe
parisc: iomap: introd
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
readq/writeq were not defined when iomap.h was included. Thus I've
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
Logan Gunthorpe (6):
iomap:
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
On 13/07/18 05:38 PM, Guenter Roeck wrote:
> This patch causes a build failure in -next when building
> ppc:corenet64_smp_defconfig.
Thanks for the report. This 64bit IO stuff is a bit of a mess. It looks
like your corner case arch and config was never covered by the kbuild
robot which has run
On 7/4/2018 11:32 AM, Andy Shevchenko wrote:
> It means it would be invariant to LE BE accessors, right?
>
> lo-hi: LE (0x0 0x4) BE (0x4 0x0)
> hi-lo: LE (0x4 0x0) BE (0x0 0x4)
Ok, well, given that this is what I implemented originally and the
argument seems a little bike-sheddy. I vote we just
On 7/4/2018 11:21 AM, Andy Shevchenko wrote:
> I think it doesn't contradict with what you are saying rather supports it.
>
> I would expect to have lo-hi and hi-lo semantics done according to the
> data flow, not to the address.
Hmm, no, I expected the opposite...
Logan
On 7/4/2018 11:16 AM, Andy Shevchenko wrote:
> On Wed, Jul 4, 2018 at 8:13 PM, Logan Gunthorpe wrote:
>> On 7/4/2018 11:10 AM, Andy Shevchenko wrote:
>>> We have an iDMA 32-bit hardware (see drivers/dma/dw/) which has an
>>> extension 64-bit registers where only one
On 7/4/2018 11:10 AM, Andy Shevchenko wrote:
> We have an iDMA 32-bit hardware (see drivers/dma/dw/) which has an
> extension 64-bit registers where only one of them has a specific bit
> to "commit" the changes written to all of them. And by some very
> unknown reason that bit is in lo part which a
On 7/4/2018 9:06 AM, Fabio Estevam wrote:
>> Logan, you mentioned the following (which unfortunately I somehow missed):
>> https://lore.kernel.org/lkml/c3f2e061-5ed1-5c74-b955-3d2bfb0da...@deltatee.com
>> The lo_hi/hi_lo functions seem to always refer to the data being written
>> or read not to the
ISA ARM
> Segment none
>
> Revert it for now.
>
> Signed-off-by: Fabio Estevam
Acked-by: Logan Gunthorpe
On 03/07/18 05:57 PM, Logan Gunthorpe wrote:
> To confirm, can you try the patch below?
Actually, scratch that: try this patch as I forgot the read side on the
previous one.
Logan
--
diff --git a/drivers/crypto/caam/regs.h b/drivers/crypto/caam/regs.h
index 5826acd9194e..796ff764dcbf 100
On 03/07/18 04:21 PM, Andy Shevchenko wrote:
> It is an explicit call to BUG().
> That's why we see wrong instruction trap.
Ok, I think I see the problem... the code is rather confusing:
Prior to the patch, IOs were BE depending on caam_little_end but if
caam_imx was set, then it wrote two LE
On 03/07/18 04:21 PM, Andy Shevchenko wrote:
> On Wed, Jul 4, 2018 at 12:20 AM, Logan Gunthorpe wrote:
>> On 7/3/2018 1:40 PM, Fabio Estevam wrote:
>
>> Also, it could be helpful if you can compile with CONFIG_DEBUG_INFO and
>> use gdb to find out what line "caam_
On 7/3/2018 1:40 PM, Fabio Estevam wrote:
>> So I have no idea what's going on here... Are we sure this is the patch
>> causing the problem? Did you bisect?
>
> Yes, I am sure that 46e4bf08f6388ba748 is the one causing the kernel boot
> issue.
Also, it could be helpful if you can compile with CO
On 7/3/2018 2:10 PM, Fabio Estevam wrote:
> On Tue, Jul 3, 2018 at 4:58 PM, Andy Shevchenko
> wrote:
>
>> By the way, is there any URL which contains dmesg out of kernel with
>> this commit reverted?
>
> Yes, here is the linux-next 20180626 dmesg (which is the last
> linux-next that does not con
On 03/07/18 12:11 PM, Fabio Estevam wrote:
> On Tue, Jul 3, 2018 at 3:09 PM, Andy Shevchenko
> wrote:
>
>> Btw, what is the environment did you use to build it?
>> and what is the environment / make variable you supply (like ARCH,
>> CROSS_COMPILE, etc)?
>
> I build it on a Dell laptop running
On 03/07/18 11:36 AM, Andy Shevchenko wrote:
>> This is now in linux-next as commit 46e4bf08f6388 and it breaks
>> booting imx6 (32-bit ARM):
>
>
>> Any ideas on how to fix this issue?
>
> Oops, first of all the header should be hi-lo instead of lo-hi.
> Does it fix it?
>
> Otherwise I didn'
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
hen iomap.h was included. Thus I've
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
Logan Gunthorpe (7):
iomap: Use non-raw io functions for io{read
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
/ written to first, followed by the upper address. Indeed the I/O
accessors in CAAM driver currently don't follow the spec, however this
is a good opportunity to fix the code.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Horia Geantă
Reviewed-by: Andy Shevchenko
Cc: Dan Douglass
Cc: Herbe
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
/ written to first, followed by the upper address. Indeed the I/O
accessors in CAAM driver currently don't follow the spec, however this
is a good opportunity to fix the code.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Horia Geantă
Reviewed-by: Andy Shevchenko
Cc: Dan Douglass
Cc: Herbe
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
s in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
Logan Gunthorpe (7):
iomap: Use non-raw io functions for io{read|write}XXbe
parisc: iomap: introduce io{read|write}64
iomap: introduce io{read|write}64_{lo_hi|hi_lo}
io-64-nonatomi
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
arlier versions were drastically different)
--
Logan Gunthorpe (7):
iomap: Use non-raw io functions for io{read|write}XXbe
parisc: iomap: introduce io{read|write}64
iomap: introduce io{read|write}64_{lo_hi|hi_lo}
io-64-nonatomic: add io{read|write}64[be]{_lo_hi|_hi_lo} macros
ntb: ntb_hw_intel:
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
/ written to first, followed by the upper address. Indeed the I/O
accessors in CAAM driver currently don't follow the spec, however this
is a good opportunity to fix the code.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Horia Geantă
Reviewed-by: Andy Shevchenko
Cc: Dan Douglass
Cc: Herbe
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
On 4/4/2018 4:38 AM, Michael Ellerman wrote:
What's the plan for getting it merged? Seems we don't have one?
Yeah, so far there is no plan. I'm not really sure who's attention I
need to get or how to get it.
Given it touches two arches and generic stuff and drivers and crypto,
it's a bit of
Add support for io{read|write}64() functions in parisc architecture.
These are pretty straightforward copies of similar functions which
make use of readq and writeq.
Also, indicate that the lo_hi and hi_lo variants of these functions
are not provided by this architecture.
Signed-off-by: Logan
Clean up the ifdefs which conditionally defined the io{read|write}64
functions in favour of the new common io-64-nonatomic-lo-hi header.
Per a nit from Andy Shevchenko, the include list is also made
alphabetical.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Jon Mason
d. Thus I've
included a patch to adjust this
- Fixed some mistakes with a couple of the defines in io-64-nonatomic*
headers
- Fixed a typo noticed by Horia.
(earlier versions were drastically different)
--
Logan Gunthorpe (9):
iomap: Use non-raw io functions for io{read|write}XXbe
pa
Now that ioread64 and iowrite64 are available in io-64-nonatomic,
we can remove the hack at the top of ntb_hw_intel.c and replace it
with an include.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Acked-by: Dave Jiang
Acked-by: Allen Hubbe
Acked-by: Jon Mason
---
drivers/ntb/hw
code did
not annotate correctly.
Signed-off-by: Logan Gunthorpe
Cc: Thomas Gleixner
Cc: Kate Stewart
Cc: Philippe Ombredanne
Cc: Greg Kroah-Hartman
Cc: Arnd Bergmann
Link:
http://lkml.kernel.org/r/cak8p3a25zqdxyay3ivv+jmsszs7f6ssgc+hdbkgs54zfvix...@mail.gmail.com
---
lib/iomap.c | 8
encouraged to use ioreadXX, et al instead
of readX[1], et al -- and mixing ioreadXX with readq is pretty ugly.
[1] LDD3: section 9.4.2
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Christoph Hellwig
Cc: Arnd Bergmann
Cc: Alan Cox
Cc: Greg Kroah-Hartman
---
include/linux/io-64
/ written to first, followed by the upper address. Indeed the I/O
accessors in CAAM driver currently don't follow the spec, however this
is a good opportunity to fix the code.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Horia Geantă
Reviewed-by: Andy Shevchenko
Cc: Dan Douglass
Cc: Herbe
Subsequent patches in this series makes use of the readq and writeq
defines in iomap.h. However, as is, they get missed on the powerpc
platform seeing the include comes before the define. This patch
moves the include down to fix this.
Signed-off-by: Logan Gunthorpe
Acked-by: Michael Ellerman
and writeq are defined. If
they are not, then the wrappers that always use non-atomic operations
from include/linux/io-64-nonatomic*.h will be used.
Signed-off-by: Logan Gunthorpe
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
Cc: Arnd Bergmann
-by: Logan Gunthorpe
Tested-by: Horia Geantă
Reviewed-by: Andy Shevchenko
Cc: Benjamin Herrenschmidt
Cc: Paul Mackerras
Cc: Michael Ellerman
---
arch/powerpc/kernel/iomap.c | 40
1 file changed, 40 insertions(+)
diff --git a/arch/powerpc/kernel/iomap.c
On 26/03/18 01:50 PM, Arnd Bergmann wrote:
> I wouldn't expect it to matter: the byte swap is almost always much
> cheaper compared to the actual bus access for the MMIO, and I
> would also guess that modern compilers can eliminate the double
> byte swap on architectures where writel() is an inli
1 - 100 of 317 matches
Mail list logo